ILO is a specialized agency of the United Nations

267th Session
Geneva, November 1996

Working Party on the Social Dimensions of the Liberalization of International Trade



Continuation of discussions concerning the
programme of work and mandate of the
Working Party:

(a) Replies to the questionnaire on the impact of
globalization and trade liberalization on the attainment
of ILO social objectives


I. Impact

II. Attitudes to trade liberalization and globalization

Changes in national legislation

III. International action and cooperation

IV. General comments, suggestions and information


1. At its 265th Session (March 1996), the Working Party on the Social Dimensions of the Liberalization of International Trade asked the Office to write to constituents in member States with a simple questionnaire requesting their views on what had happened or was expected to happen as a result of the liberalization of international trade and globalization.(1) Of particular interest to the Working Party was the impact on social standards, notably member States' ability to observe the fundamental human rights Conventions, as well as other important implications in the labour field. The Office was asked to produce a synthesis of the results for submission to the Working Party. The present paper is submitted by the Office in response to this request. The Appendix gives a statistical breakdown of the replies.

2. The questionnaire was prepared by the Office and its drafting was the subject of as much coordination with the Chairperson of the Working Party and representatives of the social partners as the limited time available permitted. Copies of the questionnaire are available separately. In order to obtain swift replies, the questionnaire was distributed to the heads of delegation of all member States of the ILO at the International Labour Conference in June 1996, with a request that they should also forward it to the most representative employers' and workers' organizations.(2) The recipients were requested to send in their replies to reach the Office by 15 September 1996. In mid-August 1996, field offices of the ILO were asked to urge constituents to reply to the questionnaire and to provide any help that might be appropriate in preparing the replies. As only some 20 replies had been received by 15 September 1996, it was decided to extend the deadline. This report is based on a total of 136 replies received at headquarters by 30 October 1996 from 128 constituents in 76 countries (appendix, table 1). From 11 of these countries replies were received from all three constituents.

3. As stressed in the introduction to the questionnaire, the purpose of the survey and what it hoped to accomplish are modest. It did not pretend to be a scientific exercise or to provide an exhaustive measurement of the impact of trade liberalization and globalization in the social field: it was only a very preliminary evaluation of their impact through the reasoned perceptions and expectations that the actors and social partners have of trade liberalization and globalization. The questionnaire also attempted to evaluate the directions that national and international actions might take to reconcile this phenomenon with the attainment of ILO social objectives. The questionnaire covered the period 1985-95.

4. Mention must be made at the outset of some of the more important limitations of the exercise, several of which were also pointed out by some constituents. The first and perhaps most important limitation was that it was often difficult, even with the best scientific tools at hand, to distinguish the impact of trade liberalization and globalization from that of other factors such as macroeconomic and labour market conditions, market reforms and public policies. In addition, statistical information on some of the variables in the questionnaire may not be available in many countries, particularly developing countries. Secondly, the precise scope of what is understood by the term "globalization" may vary from one respondent to another. While the questionnaire defined globalization as economic interdependence through trade in goods, services and capital flows, it was not entirely clear, for example, whether the impact of market reforms should or should not be subsumed under that of globalization, since it may legitimately be argued that such reforms were often part, precisely, of efforts to facilitate growing integration into the world economy. Thirdly, a certain ambiguity may also be present in some questions and answers as to whether they related to respondents' own positions or to their understanding of empirical evidence available to them. Fourthly, when answers suggested that trade liberalization and globalization had no effect on a particular variable, it was not always clear whether what was meant was that these processes indeed had no effect on the variable of interest, or that there was no substantial trade liberalization or globalization in the first place. Finally, the perception of the particular representative organization answering the questionnaire may differ from that of another in the same constituency.

5. A principal aim of the survey was to provide a fairly comprehensive and representative picture of governments', employers' and workers' reasoned perceptions, expectations, concerns and projects regarding the phenomenon of globalization and trade liberalization. The extent to which the sample of replies is representative of ILO constituents may be assessed from the statistics in table 2: 44 per cent of the replies are from governments, 23 per cent from employers and 33 per cent from workers.(3) The overall response rate (the percentage of ILO constituents replying) was 25 per cent. There are, however, large differences across various categories of constituents: while 33 per cent of governments and 24 per cent of workers' organizations had replied by 30 October 1996, the response rate for employers was only 17 per cent. The differences are even larger across various types of economies and regions.(4) Industrialized countries had the highest rate of response (53 per cent) and the least developed countries the lowest (9 per cent). The rates for developing countries as a whole (19 per cent) and for transition economies (23 per cent) are somewhat below the average of 25 per cent for the entire sample. From a regional perspective, Europe had the highest rate of response at 35 per cent, and Africa the lowest at 13 per cent. Employers' organizations, the least developed countries and Africa are hence significantly under-represented in the sample of replies, while governments, the industrialized countries and Europe are relatively over-represented.(5) In view of the modest rate of response overall, the non-random nature of the sample of replies and its potential biases, considerable caution is called for in interpreting the findings reported below.

6. In presenting the results, responses have been tabulated by the constituent responding (governments, employers and workers), the type of the economy from which the reply comes (developing, transition and industrialized), as well as for the entire sample.(6) These criteria are intended to reflect categories in which potential differences of views on the implications of trade liberalization and globalization have often tended to be expressed. A more detailed classification of results (for example, by the category of constituent together with the type of economy), though used occasionally, is constrained by the limited size of the sample. The scope of the analysis is also more limited than originally intended due to the need to extend the deadline for the receipt of replies. It is nevertheless hoped that the findings reported below will be of use to the Working Party. The section headings follow those in the questionnaire.

* * *

I. Impact

A. Background and trends

7. The questionnaire requested certain background data and information on employment and its breakdown by gender and sector in the economy as a whole as well as in the main export sectors. The respondents were also requested to provide information as to whether, over the past ten years, any sectors had changed from being substantial net exporters to being substantial net importers, or vice versa. For various reasons it unfortunately proved difficult to utilize this information systematically: the extreme heterogeneity in the level of disaggregation of sectors, the relatively small size of the sample, which did not permit the classification of replies to other questions by reference to the background information provided, or simply the inaccuracies in the data. A few observations, however, may be made on the basis of the information provided. First, in virtually all developing countries for which governments had provided total employment figures for both 1985 and 1995 (or years close to them), employment had increased, from a few percentage points in some countries to over 100 per cent in Kenya. The increase was of the order of 30 to 40 per cent in most cases. The changes in industrialized countries were more modest and in both directions, varying from -15 to +15 per cent over the last decade. By contrast, total employment had declined in all European transition countries in the sample by roughly 10 to 20 per cent over the same period.

8. Secondly, the share of women in employment in 1995 (or a year near to it) was mostly in the range of 20 to 40 per cent in developing countries, around 50 per cent in transition economies, and from 35 to 48 per cent in the industrialized countries in the sample. This share increased somewhat (in most cases up to 4 or 5 percentage points) in a large majority of countries in the sample. The most notable declines were in four transition economies in Europe (2 to 5 percentage points). The share of women in total employment in the main export sectors varied much more across countries. In 1995 (or a year close to it), this share ranged from 10 per cent (Guyana) to 90 per cent (Lesotho) in the developing countries that provided the requested data, from 40 to 65 per cent in the European transition countries, and from about 25 per cent in several industrialized countries to as much as 70 per cent in New Zealand.

9. Thirdly, as far as a change in the trading position of various sectors is concerned, 21 per cent of those answering the relevant question stated that some sectors of their economy had gone from being substantial net exporters to being substantial net importers, with 18 per cent indicating the opposite. The changes occurred in all types of economies and sectors. To give but two illustrations, in China, crude oil, wheat, maize and rice had turned from substantial net exporting sectors to substantial net importers, whereas in the case of household electrical appliances, watches and telecommunication equipment the opposite had occurred. In the case of Finland, to take an example from an industrialized country, textiles, clothing, shoes and leather had become net importers, whereas metal and engineering industry products had turned into substantial net exporters.

B. Employment and wages

10. Constituents were asked for their views on the effects of trade liberalization and globalization on the level of employment, real wage rates and wage disparity, both over the past ten years as well as in the future. Table 3 provides a summary of the results.(7) The figures in the table represent the percentage distribution of answers (ignoring non-response) for each category of constituents (or, in the last column, for all of them combined). The table also reports "scores" which are designed to facilitate inter-group comparison by providing an overall measure of the views of those responding to a particular question. They are calculated as follows. A score of +1 is assigned to the first of the three multiple-choice answers (in this case "increase"), -1 to the second, opposite answer ("decrease"), and 0 to the last option ("no substantial effect"). The scores reported in the table are mean scores (in percentages) for the responses of all those in a particular category who answered the relevant question. Mean scores represent, in effect, the percentage of replies choosing the first answer ("increase") minus the percentage choosing the second, opposite answer ("decrease").(8)

11. Considering the effect on employment first (in net terms -- jobs created minus jobs lost), it may be seen that while a quarter of replies from governments and nearly one-half of those from employers saw no substantial effect during the past ten years, the remaining replies are evenly split between "increases" and "decreases". This even split, representing a middle-of-the-road position, is reflected in the low (near-zero) "scores" for these two groups of constituents. By contrast, workers' replies suggest a broadly negative assessment (a fairly high negative score) of the effect of trade liberalization and globalization on employment, with 56 per cent considering that these processes destroyed more jobs than they created over the past ten years, 28 per cent holding the opposite view, and 16 per cent seeing no substantial effect. The contrast between governments' and employers' perceptions on the one hand, and that of workers' on the other, also broadly holds for the effect on real wage rates in the past. It in fact sharpens in regard to the expectations about the future impact of trade liberalization and globalization on employment and real wages. Thus, while workers' replies are at least as pessimistic about the future as they were about the past, employers, and especially governments, move from their middle-of-the-road assessment of past effects to considerable optimism about the future effects. At least 60 per cent of government replies expected these processes to increase net employment and real wages in the future. The majority of workers' replies, however, expected the opposite.(9)

12. The classification of results by type of economy suggests that, unlike replies from developing and industrialized countries, those from transition economies have a negative assessment of the impact, over the past ten years, of trade liberalization and globalization on employment and real wages. They do, however, suggest the expectation of considerable improvements in the future; those from developing countries do so, but not so strongly; while those from industrialized countries do not.

13. Turning now to the perceptions of effects on wage disparities, there is a widely shared view that trade liberalization and globalization have failed to have an equalizing effect on wages. As may be seen in table 3, only about 10 per cent of the replies, whatever the classification used, suggested that these processes had reduced wage disparities over the past ten years. There are some nuances, however: while the perceptions of governments and workers are practically identical -- with a majority of each stating that disparities had widened as a result of these processes -- two-thirds of employers' replies detected no substantial effect. As regards the future, both governments and employers are, on the whole, of the view that these processes will have somewhat less of a tendency to exacerbate wage disparities in the years to come, but workers do not seem to share this expectation.

14. Respondents were requested to provide their assessments of the main reasons for their expectations about the future effect of trade liberalization and globalization on employment. The expectation that these processes are likely to improve employment prospects appears to be rooted, as indicated in many replies, in the belief that they lead to the opening up of world markets to exports, increases in output and hence greater employment opportunities. Several respondents, however, pointed out that national policies should be able to adapt to take advantage of the new opportunities created by trade liberalization (Republic of Korea, G; United Kingdom, E and G). Among other reasons cited are the increase in domestic or foreign direct investment (Cambodia, G; Honduras, G; India, E), the reduction in transaction costs due to the lowering of tariff barriers and increased transparency in markets (Netherlands, G) and cheap labour as an attraction for international investors (Lesotho, G). Some were also hopeful that, following the inevitable costs of structural adjustment, their national economies were now in a better position to create jobs (Egypt, G; Switzerland, G). A few other respondents based their optimistic expectations on the fact that their countries had already put in place, or were likely to do so in the near future, more coherent policies to encourage investment (Mali, G) and the acquisition of skills (Guyana, G).

15. On the other hand, those who were not so hopeful about the impact on employment in the future pointed to such factors as the emergence of free markets in transition economies (Lithuania, E); privatization (Botswana, W); economic uncertainty and political instability; the deflationary policies of the OECD countries (Italy, W); inflexibility in labour markets and high labour costs; the absence of active labour market policies or industrial policy (Argentina, W); automation to reduce costs and compete more effectively (Luxembourg, W; Trinidad and Tobago, G and W); rapid technological innovation; inadequate availability of skilled labour to facilitate the timely introduction of modern technology; severe competition and cheap imports from major new producers of certain types of commodities (Barbados, E; France, W); the introduction of EPZs (Botswana, W); and rapid technological innovation, among a number of other factors.

C. Freedom of association and union membership

16. This section of the questionnaire began by requesting respondents to provide data on the rate of union membership (approximate percentage of labour force belonging to trade unions). The results are presented in table 4. It may be seen that in some cases the figures reported by the constituents in the same country differed, sometimes substantially. For example, the 1985 rate for Trinidad and Tobago was put at 14 per cent by the Government, 21 per cent by employers and 30 per cent by workers, although all three agreed that it had declined over the past ten years. Despite such discrepancies, table 4 shows that, while the rate of union membership did not change much between 1985 and 1995 (or the years nearest to them) in a number of countries, in many others it did, declining at least twice as often as it increased (in cases where there was a change of at least 5 percentage points between the two periods). There were fairly significant increases in such countries as Mozambique and South Africa, and decreases in, for example, Argentina and New Zealand. The most dramatic declines, however, occurred in the transition countries (except China) where trade unions often encompassed almost all workers in the initial period (1985), but may now cover no more than a third of workers.(10)

17. Other results relating to this section are summarized in table 5. Union membership rates in the main export sectors do not appear to be systematically higher or lower than those in other sectors of the economy. About half of the replies saw no significant difference, and where a difference was reported, the answer was as likely to be "higher" as "lower". Many replies, however, indicated that unionization rates had been on the decline both in the main export sectors and in sectors facing import competition. The decline in the main export sectors appears to have been less frequent, in relative terms, in developing countries than in industrialized countries, and especially in transition economies where it was generalized.

18. About half of the replies from industrialized countries indicated no change in the proportion of workers covered by collective agreements during the past ten years, either in the economy as a whole or in the main export sectors. Most replies from developing countries and transition economies, however, indicated that there was indeed a change, although it was not systematic. Where there was a change, an interesting difference emerges between governments, on the one hand, and employers and workers on the other: while the former suggested that the proportion of workers covered by collective agreements had more often increased than decreased, the latter believed that it had more often declined.

19. Most replies, particularly those from governments and employers, expressed the view that, overall, trade liberalization and globalization had had no effect on the exercise of freedom of association or the right to engage in collective bargaining. It is interesting to note that nearly half of the replies from workers echoed this view as well. However, when an effect is indeed perceived, there is a sharp contrast between workers -- who consider that these processes have led to a limitation of such rights -- and governments and employers, who tend to feel the opposite. The reasons provided by the proponents of "extension" often have to do with the emergence of a market economy, privatization and the spread of democracy, all leading to the strengthening of social partners and opportunities for the exercise of freedom of association and collective bargaining (Estonia, G; Lithuania, E; Romania, G). In the more specific case of South Africa, entry into world trade is said to have required improvements in trade union rights (South Africa, E). The view has also been expressed (New Zealand, E) that accepting freedom of association includes the right not to associate, and the change in numbers not covered by collective agreements reflects the exercise of the right to choose and to opt for coverage under an individual employment contract, as well as a move to enterprise bargaining. The proponents of "limitation" also in some cases invoked the same reason to back up their assessments. For example, privatization is cited as entailing a weakening of the role of unions through reductions in the number of workers (Trinidad and Tobago, G). Most assessments, however, link the limitation of the exercise of such rights to such factors as the spread of MNEs (Luxembourg, W), the curtailment of union activities in EPZs (Jamaica, E), the dismantling of labour codes (Burkina Faso, W), and also more severe competition compelling enterprises "to lessen costs in order to compete with countries that do not respect the social clause" (Argentina, W).

D. Equality of opportunity and treatment

20. Where the percentage of women in the main export sectors was different from that in other sectors (more than one-third of replies suggested that it was not), most replies indicated that it was lower (table 6). Over the past ten years, the share of women workers in the main export sectors is more often said to have risen than fallen, with about half of the replies from governments, employers and workers suggesting no change. It is in developing countries that the share of women appears to have been rising most generally. As regards the perception of the effect of trade liberalization and globalization on equality of opportunity and treatment for workers over the past ten years, workers' replies are far more pessimistic than those from governments and employers, with more than 40 per cent of them considering this effect to have been negative. Positive effects are said to include the spread of employment opportunities, the contribution to higher productivity, improved equality of opportunity and treatment, greater integration of women in the economy and better recognition of their contribution, equal rights in negotiations, freedom of association and the right to strike. Among the negative effects cited are job losses, which may have been concentrated in sectors with a high proportion of women workers, growing poverty, the growth of precarious forms of employment, and the particularly difficult situation of older and unskilled workers seeking employment. The reasons advanced for a positive effect of trade liberalization and globalization on equality of opportunity and treatment for workers include the increase in exports whose beneficial effects on employment and greater economic dynamism offer more opportunities for women. On the other hand, the unbridled pursuit of profit and the insensitivity of the globalized economy to social needs (Peru, W) figure among the reasons given by the respondents for a negative effect.

E. Child labour

21. This section of the questionnaire began by requesting data on the approximate percentage of children of primary school age enrolled in school in 1985 and 1995, separately for boys and girls as well as for both combined. Almost all replies from industrialized countries put the enrolment ratios at 99 or 100 per cent.(11) The figures ranged from 90 to over 100 per cent (where children above primary school age were also attending such schools) in the few transition countries that responded to the question.(12) However, the situation in developing countries is of course much more diverse: while in some countries (such as Singapore), all children went to school, in some African countries only about a third of them did so. In all developing countries answering the question (fewer than 20 did), however, the enrolment ratio had increased over time. The enrolment ratio for girls was in some cases substantially lower than that for boys, but not always. In the case of Lesotho, while nearly 80 per cent of girls went to school in 1993, the corresponding ratio for boys was only about two-thirds.

22. The questionnaire defined child labour as labour performed by children under the legal age of admission to employment in the country concerned. In nine cases, government replies stated that the reported violations of child labour had increased over the past ten years. In four countries (notably two in Eastern Europe), this was thought attributable to a rise in the incidence of child labour, whereas in one greater compliance was believed to be the reason. No reason was given for the increase in reported violations in the remaining four countries. On the other hand, seven governments indicated a decrease in reported violations and attributed this to greater compliance. In 19 countries in the sample, the number of reported violations remained more or less the same during the past ten years.(13)

23. As may be seen in table 7, 42 per cent of the replies reported that child labour was absent in the economy as a whole. Fifty-three per cent of them, including several from industrialized countries, indicated that child labour was present to a "small extent" in their countries. Whatever the breakdown of the replies, however, the extent of child labour appears to be lower in the main export sectors as compared to the economy as a whole. In five developing countries, child labour was reported to be present to a "large extent" in the economy as a whole. The same is true of the main export sectors in as many (though not always the same) countries. About one-quarter of the replies provided an estimate of the proportion of child labour in the country that is found in the main export sectors, with most putting it at less than 1 per cent. But four replies suggested that this proportion was at least 20 per cent.

24. Most replies answering the relevant question indicated that the extent of child labour in the main export sectors had either not changed or had declined during the past ten years. Some 33 per cent of the replies from workers, however, suggested the opposite (as compared to 12 per cent of those from governments and none from employers). The respondents were asked to give their assessment of the main reasons for a change or lack thereof in the extent of child labour in the main export sectors over the past ten years. A wide range of factors were cited: the availability of a large pool of cheap labour (Lesotho, G), high unemployment (Dominica, E; Pakistan, W), high unionization in the main export sectors (South Africa, E) and traditional family values (Dominica, E) were mentioned as reasons why child labour was limited and not growing. An increase was often said to be associated with a decline in living standards and a rise in poverty, whether or not resulting from structural adjustment, the growth of the informal sector (Bolivia, E), lack of adequate monitoring (Fiji, W) and laxity in the application of labour laws prohibiting child labour. Significantly, the changing attitude in the society was also mentioned, along with a decline in living standards, in a workers' reply from an East European country as a reason for the increase in child labour in the main export sectors. Where a decline was reported in the extent of child labour in the main export sectors, this was usually explained by reference to the decline in poverty, an increase in schooling (Italy, G; Republic of Korea, G) and the facilitation of primary education by, for example, the supply of free textbooks, midday meals and allowances for clothing (Sri Lanka, E), strict monitoring and international pressure (Pakistan, W), as well as awareness campaigns (Bangladesh, G; Mauritius, G) and the effective application of relevant labour laws.

F. Working conditions

25. As table 8 shows, one-half of the replies stated that occupational accident/illness rates are about the same in the main export sectors as compared to other comparable sectors producing for the domestic market; the other half was about evenly divided between "higher" or "lower". As far as the change over the past ten years is concerned, a majority of governments and employers considered that the occupational accident/illness rate had fallen in the economy as a whole as well as in the main export sectors. Nearly one-half of the workers' replies, however, indicated the opposite. On the whole, the situation in industrialized countries appears to have improved considerably over the past ten years, but in transition economies the replies indicating a deterioration were three to four times more numerous than those suggesting an improvement.

G. Social security

26. The questionnaire enquired about constituents' perceptions of the possible impact of trade liberalization and globalization on certain aspects of the social security systems in their countries over the past ten years. Leaving aside the difficulty of attributing any possible changes to these processes, the most meaningful differences arise when the replies are disaggregated by type of economy (table 9). An overwhelming proportion of replies from industrialized countries (four-fifths) reported no effect as far as coverage was concerned, both in terms of the percentage and the categories of workers covered. About a third of the replies, however, indicated a decrease in the benefits provided to the workers covered. Some 40 per cent of the replies suggested that the net financial sustainability of the system (i.e. the difference between change in resources compared to change in benefit liabilities) had deteriorated during the past ten years, with the reverse being stated in relatively few instances. Furthermore, as many as three-quarters of the replies reported no changes in the legal status or operation of the system.

27. The most deleterious effects on social security systems are observed in transition economies. Here, a substantial proportion of the replies suggested a decrease in coverage and in the benefits provided. Most replies, furthermore, stated that the net financial sustainability of the system had suffered. A vast majority of them also indicated some change in the legal status and operation of the system. The picture that emerges for the developing countries is on the whole more encouraging. While about half of the replies reported no effect in response to the relevant questions, both coverage and benefits were more often reported to have increased than decreased, and the net financial sustainability of the system had as often improved as deteriorated.

H. Export processing zones (EPZs)(14)

28. About two-thirds of the replies did not include answers to the questions in this section of the questionnaire, perhaps because EPZs are absent in the countries concerned and are not envisaged in the near future. The share of EPZs in total exports of the countries having them (about one-fifth of the 76 countries in the sample, mostly developing and a few transition countries) varies from 2 per cent in Kenya and the Republic of Korea to 50 per cent in Belarus, and as much as 63 per cent in Mauritius. The number of people working in the EPZs also varies substantially, from a few hundred in several countries to more than 80,000 in Mauritius and Sri Lanka in 1995. Most such workers are women. Women accounted for some 70 per cent of the total in Bangladesh, Colombia and Mauritius, and more than 80 per cent in Honduras and Sri Lanka, to mention only cases with a sizeable EPZ workforce. As to the future, three-quarters of those answering the question said that they expected the level of employment in the EPZs to increase in the next ten years. The replies to the question as to whether the terms and conditions of employment in the EPZs were the same as or different from those in comparable sectors in the rest of the economy drew sharply different answers. While most governments and half of employers suggested that these were the same, almost all workers' replies indicated that they were different.

29. As explained by the respondents, the terms and conditions of employment in EPZs could differ from those in comparable sectors in respect to wages and fringe benefits, duration of contract, hours of work, welfare facilities and the application of the labour code and of international labour standards, among other factors. For some constituents, employment conditions in the EPZs were better (Malaysia, E; Sri Lanka, E; Ukraine, G). For others, particularly among workers, this was not so. They pointed to the lower wages and benefits in some cases (Jamaica, E; Senegal, W), the absence of social security, old-age benefits or profit sharing (Pakistan, W), temporary rather than permanent contracts (Bolivia, E), and the non-application of labour laws and standards in such areas as freedom of association and the right to engage in collective bargaining (Cameroon, W; Pakistan, W; Trinidad and Tobago, W). The most often cited reason for this was the need to attract investment under fierce competition (Argentina, W; Senegal, W; among others).

II. Attitudes to trade liberalization and globalization:
Changes in national legislation

30. Respondents were requested to specify whether, in the context of the global economy, various factors act as advantages or disadvantages for their countries. In order to simplify the interpretation of the results, the five options for the response relating to each factor were assigned scores (2 for a big advantage, 1 for a small advantage, 0 for neutral, -1 for a small disadvantage, and -2 for a big disadvantage). Mean scores for various factors and groups are reported in table 10. It may be observed that governments and workers rank the skills of the labour force as, on the whole, the most advantageous of the options given. Workers' replies do not rank wage costs as particularly advantageous and government replies agree. Employers' replies, on the other hand, consider wage costs as acting as a disadvantage for their countries, as they do social security costs. They rank the labour relations climate as providing the most advantage of all six factors, but workers appear to be less convinced.

31. In industrialized countries, wage costs and social security costs were regarded by the respondents as somewhat of a disadvantage, whereas the skills of the labour force, the labour relations climate, labour productivity, and the capacity to innovate or adapt were regarded as advantageous, in decreasing order of importance. The low wage costs in transition economies make this factor the most advantageous to them, followed by the skills of the labour force. None of the factors was thought particularly disadvantageous. From the perspective of respondents in developing countries, all factors appeared to provide at least some advantage, except perhaps wage costs which were perceived as being largely neutral. Among additional factors mentioned as a big advantage were land reform (Pakistan, W), geographic location (Senegal, W) and political stability permitting low interest rates (Switzerland, G), and as a big disadvantage, quality of products (Romania, G; Mauritius, G), research and development (Mauritius, G) and high public debt (Italy, G).

32. To take better advantage of the opportunities created by globalization and trade liberalization, the replies confirm that many countries have attempted to pave the way by modifying national legislation in a variety of subject-areas. Respondents were requested to provide their assessment as to whether, to meet the perceived needs of trade liberalization and globalization, their national legislation had been substantially modified in the past ten years, or was expected to change in the next few years, in respect to each of a number of subject-areas. Table 11 provides a summary of the answers. It may be seen that, over the past ten years, national legislation appears to have been substantially modified most commonly with the purpose of promoting exports, attracting foreign investment and improving training and retraining. Over 70 per cent of those responding to the relevant questions indicated that this was the case in their countries. Changes in legislation in regard to the establishment or promotion of EPZs, labour subcontracting, statutory minimum wage, terms of engagement, working time, termination of employment and other changes to increase labour market flexibility were also mentioned by around 40 to 50 per cent of respondents. The least common type of change concerned special schemes for workers unemployed as a result of the effects of international trade and other measures to mitigate the possible negative effects of globalization. As far as the next few years are concerned, in all the subject-areas a higher percentage of respondents believed that more changes were in the offing than in the past.

33. Nearly three-quarters of the respondents expected to see legislative changes in a variety of additional areas to increase labour market flexibility. These include such areas as wage reform towards a more flexible system (Singapore, G), private employment agencies (Italy, E), labour contract systems (Pakistan, W), and industrial and vocational training (Mauritius, G). Legislative changes were also introduced, or are expected to be introduced in the near future, in relation to reconversion (Senegal, W), training in new skills and entrepreneurship development (Mauritius, G), and retraining, temporary employment, and allowances (Bulgaria, G) in an attempt to mitigate the possible negative effects of international trade and globalization.

34. Respondents were asked for their opinion on the possible effect of trade liberalization and globalization on the means of action available to their governments to promote the social goals of the ILO. Roughly half of them, in total or in individual categories, did not see any effect (table 12). Of the rest, most workers' replies considered that the means of action had been reduced while most employers' replies expressed the opposite point of view. Where these processes were said to have had an enhancing effect, this effect sometimes operated through the stimulation of economic growth arising from the expansion of trading opportunities and/or larger investment (Oman, G; New Zealand, E). A stronger economy would clearly make it easier to maintain and enhance social goals (Finland, E). In some instances mention was made of the contribution of these processes to the greater acceptance of tripartism (Cameroon, W). In South Africa, workers' representatives stressed that their country's return to international markets had facilitated the promotion of labour standards. In Colombia employers cited the creation of the multidisciplinary teams as having facilitated greater contact between employers and employees. On the other hand, where the effect was believed to have been to reduce the means of action available to the government, the most frequently noted mechanism or reason was, precisely, the privatization and reduction in the role of the State in influencing the socio-economic landscape (Senegal, W; Sweden, W), itself induced by liberalization, adjustment and globalization. With less resources available to the State, social goals were not afforded as much priority as before (Finland, W). Rising unemployment, wage erosion, the removal of subsidies, the privatization of health and education services, and labour market deregulation were also mentioned by some of the respondents.

III. International action and cooperation

35. As may be seen in table 13, overall, just over 60 per cent of the respondents did not see much of an effect of trade liberalization and globalization on their countries' ability to ratify and apply ILO Conventions on fundamental workers' rights(15) or on other labour standards. The proportion was somewhat higher for the replies from industrialized countries than for those from developing and transition economies. However, most other replies from the industrialized world were of the view that it had become more difficult for their countries to ratify and apply ILO Conventions as a result of trade liberalization and globalization. Interestingly, hardly any replies from transition economies suggested that it had become more difficult to do so: nearly half of all replies found that it had actually become easier. The developing country replies fall somewhere in between: while the majority view is that there has been no effect, the rest are divided equally between those who find that this ability has been reduced and those who hold the contrary view.

36. However, while relatively few governments and employers indicated that it had become more difficult to do so in the case of fundamental workers' rights, a third of the replies from workers suggested that in their view this was the case. This was attributed by some workers' organizations to the constraints associated with the implementation of structural adjustment policies (Burkina Faso), to the need to attract foreign investment with the incentive of low labour costs (Senegal and Pakistan), and to strengthened resistance on the part of employers (Finland). Additional reasons had to do with deregulation, the presumed adverse impact on competitiveness as well as, in the cases of Denmark and Sweden (workers), membership of the European Union and, in that of France, a feeling permeating the Government as to the desirability of a break in standard setting. The "rigid interpretation" by the ILO itself of what constitutes workers' rights was also cited by the New Zealand employers as a constraint on the ability to ratify and apply the Conventions on fundamental workers' rights. On the other hand, where this ability was said to have improved, this was attributed to such reasons as the greater use being made by employers of ILO Conventions as guidelines in the policy making process (Trinidad and Tobago, E), a new prevailing spirit of dialogue and consensus-seeking (Colombia, E), democracy brought about by the market economy (Estonia, W), or as a consequence of harmonization with the expectations of the European Union (Hungary, W). The employers from Trinidad and Tobago also noted that occupational health and safety standards were being increasingly adhered to as companies vied for certification by the International Standards Organization (ISO) as a key to accessing foreign markets.

37. There is a very large measure of agreement among respondents (84 per cent) that trade liberalization and globalization call for universal recognition and the application of fundamental workers' rights. Workers, as might be expected, took this position almost unanimously (95 per cent of replies), but 82 per cent of governments and 70 per cent of employers' organizations responding did the same. The respondents were asked to specify which Conventions. By far the most frequently and specifically mentioned ones were, in decreasing order of frequency: the Freedom of Association and Protection of the Right to Organise Convention, 1948 (No. 87) (approximately 50 times); the Right to Organise and Collective Bargaining Convention, 1949 (No. 98) (40 times); the Forced Labour Convention, 1930 (No. 29), the Abolition of Forced Labour Convention, 1957 (No. 105), the Discrimination (Employment and Occupation) Convention, 1958 (No. 111), and the Minimum Age Convention, 1973 (No. 138) (30 times each); and the Equal Remuneration Convention, 1951 (No. 100) (20 times). A few others were also specifically mentioned, though only rarely, such as the Employment Policy Convention, 1964 (No. 122), the Workers' Representatives Convention, 1971 (No. 135), and the Rural Workers' Organisations Convention, 1975 (No. 141).

38. Views were somewhat more diverse in regard to other labour standards. While more than 80 per cent of workers considered that trade liberalization and globalization also call for the universal recognition and application of other labour standards, governments and employers are in two minds about the matter, with about half of them not sharing this view. The same is true of replies from transition economies and industrialized countries, which were more or less split evenly for and against, although 70 per cent of replies from developing countries answered in the affirmative. The question asking respondents to specify which Conventions merited universal recognition and application drew a wide range of responses, with frequent reference to the seven already mentioned above in connection with fundamental workers' rights. Others mentioned included those relating to occupational safety and health, minimum wage, conditions of work and social security.

39. The 84 per cent of the respondents who considered that trade liberalization and globalization call for the universal recognition and application of fundamental workers' rights were further asked to give their views as to whether this aim should be pursued within the framework of the ILO, elsewhere, or not be pursued at all. Very few replies (only 2 per cent) suggested that the matter should be dropped altogether or that it should be pursued elsewhere than the ILO (3 per cent). A majority of replies (69 per cent) opted for the framework of the ILO but many (26 per cent) also favoured pursuing the matter in the framework of the ILO as well as elsewhere, although such a combined option was not in fact included in the questionnaire. However, less than a third of the replies provided specific suggestions on where elsewhere might be. The most common response (mentioned nearly 20 times) was the World Trade Organization, although those suggesting this option were mostly workers' organizations, together with a couple of governments (Russian Federation, but also, though more cautiously, Finland, which suggested that WTO links be studied, and Switzerland, which felt that informal consultations within the WTO would contribute to the process). The employers from South Africa also cited the WTO among several other fora. Among other fora mentioned in the replies, though much less frequently, are the Organisation for Economic Cooperation and Development (OECD), the European Union, the United Nations, the United Nations Conference on Trade and Development (UNCTAD) and possibly other international fora, international financial institutions, "a new body with fewer shortcomings than the ILO and which is not run solely by the developed countries" (Spain, W), trade agreements and regional institutions (Organisation of African Unity), and also through national laws, regulations and decisions governing fundamental workers' rights (Egypt, G).

40. A large number of respondents indicated that they expected more use to be made in the future of the four types of measures mentioned in the questionnaire to promote respect for international labour standards. Most widely expected are codes of conduct for enterprises or industries applied on a voluntary basis (three-quarters or more of the replies, except in the case of transition economies) and the development of a labelling system for goods and services produced in accordance with fundamental workers' rights (two-thirds or more of the replies, except in the case of employers). The next most widely expected measure were boycotts of imported products by trade unions, consumers or other non-governmental groups (54 per cent of replies), and the least expected is the extraterritorial application of national laws (42 per cent). In all four cases, a higher percentage of workers than of governments or employers expected greater use to be made of such measures in the future. The most sceptical are the replies from transition economies, with at least half of them not expecting greater use of such measures in the future except in the case of the development of a labelling system.

41. Respondents were requested to specify which, if any, of the four given options they would welcome to promote respect for international labour standards. Workers and developing countries were more responsive: nearly half of the constituents in these categories favoured at least one of the options (table 14). The corresponding proportions for other groupings were around 30 per cent. Among those specifying their preferences, the most favoured measure is the use of codes of conduct for enterprises and industries applied on a voluntary basis (39 replies in favour). The next most favoured is the development of a labelling system (30 in favour). The other two possibilities -- extraterritorial application of national laws and boycotts of imported goods and services -- drew fewer favourable responses (20 each). Workers' replies suggested that they favoured all four options more or less equally, except perhaps the extraterritorial application of national laws, which received somewhat less support than the other options. The replies from other groupings of respondents, however, were generally more in favour of codes of conduct and a labelling system than the extraterritorial application of national laws or boycotts. No employers' reply welcomed boycotts. Fifteen replies from developing countries (eight from workers, five from governments and two from employers) welcomed the extraterritorial application of national laws.

IV. General comments, suggestions and information

Suggestions for future research

42. About half of the replies contained suggestions for possible future research or indicated support for proposals that have already been put forward at previous meetings of the Working Party. Among the latter were many replies from workers expressing support for the proposals already made by the Workers' group of the Governing Body in this regard, namely, national reviews by the ILO of the respect of fundamental workers' rights by member States, a review of the effectiveness of ILO machinery to support fundamental workers' rights, a study of the use of labour market incentives to attract foreign direct investment, and a consideration of the links between respect for labour standards, productivity and economic development.

43. A recurring theme for research was the need for more detailed empirical analyses of the impact of trade liberalization and globalization on the social objectives of the ILO. In this connection, several respondents expressed support for the country studies under consideration. Some constituents recommended analyses of the social effects of the policies of the IMF, World Bank and UNDP (San Marino, E and G), and of the relationship between globalization and unemployment (Senegal, W). Narrower in focus are the more specific calls for an analysis of the impact of globalization and trade liberalization on urban centres and the informal sector (Ghana, G), as well as on industries in developing countries with small domestic markets (Sri Lanka, G). It was also proposed that a system of indicators should be developed to assess regularly the extent of exposure to globalization and trade liberalization and its impact on individual countries (Romania, G).

44. Other studies recommended related to specific dimensions of trade liberalization and globalization or to particular contexts. These include analyses of such topics as the unique situation of transition economies (Hungary, G) and the presence of multinational enterprises in them (Romania, G), employment in export sectors (direct and indirect effects) (Argentina, G), the movement of labour from one country to another taking into consideration the educational levels of workers (Panama, G), and the advantages enjoyed by developed countries arising from globalization (Sri Lanka, E). Other proposals concern the relationship between trade liberalization and labour standards (Sweden, G), and global developments in productivity and competitivity in different countries (Finland, G). Comparative studies on union rights and freedom of organization in multinational enterprises (Hungary, W) and on respect for fundamental workers' rights in EPZs (Canada, G; Finland, W; Italy, W) were also suggested.

45. Finally, several suggestions related to this type of survey itself. A follow-up questionnaire was proposed, to be sent in about five years, so as to improve understanding of the real impact of trade liberalization and globalization (India, E), particularly on developing countries (Trinidad and Tobago, E). Other proposals in this vein recommended sending appropriate questionnaires to multinational corporations (Guyana, G), and to individual trade unions or affiliates of trade union confederations (Lesotho, W) to seek their views.

Other measures and other fora

46. A number of respondents put forward suggestions on other measures and other fora through which the issue of the impact of trade liberalization and globalization on the attainment of ILO social objectives may be addressed. Among the measures proposed are: the inclusion of a social dimension in all multilateral development assistance (Sweden, W), discussion in the Governing Body's Committee on Legal Issues and International Labour Standards of the extension of supervisory procedures concerning core labour standards (Netherlands, G), the promotion of specific labour standards through coordinated programmes of technical cooperation and development assistance, conducting periodic surveys (Bangladesh, G and E), and holding discussions (United Kingdom, G) and seminars (Colombia, W; Venezuela, W) on topics related to the issue. Among other fora mentioned by the respondents are the WTO, OECD, UNCTAD and international financial institutions. The involvement of the WTO, however, was ruled out by the employers from New Zealand, who believe that the issue should be retained within the Working Party.

Ways of asserting the ILO's role

47. Constituents were asked for their views on how the ILO should assert its role and achieve its objectives on this issue within the wider international community. The suggestions may be grouped under the following main headings:

  1. Strengthening of ILO's standards-related action: Active promotion and implementation of international labour standards, in particular core standards, are recommended by some constituents as a means of asserting the ILO's role (Kenya, G; Netherlands, G; Romania, G; Italy, E). In this connection, it was also proposed that the ILO should seek international consensus on what the core labour standards are and communicate this definition to the international community (Canada, G). The strengthening of the supervisory machinery was also recommended by several constituents (Botswana, W; Panama, G; South Africa, E). The Government of the United Kingdom suggested redirecting the focus of ILO standard-setting activities towards the revision and modernization of existing standards, as well as the development of a new Convention to tackle the exploitation of child labour.
  2. Strengthening of the ILO's analytical capacity: Another approach suggested is through continuous and objective research, the dissemination of information, and the evolution of the ILO as a "centre of excellence" on the subject (Chile, G; United Kingdom, G; Netherlands, G; Switzerland, E; Germany, E; Dominica, E). A conference on trade liberalization and globalization was also suggested (Trinidad and Tobago, G).
  3. Outreach: Suggestions under this heading include the promotion of tripartism (Senegal, W) and consensus building (Trinidad and Tobago, W), technical cooperation on labour and social matters (Switzerland, E; France, W; Sri Lanka, E; Malaysia, E), awareness raising and education (Lesotho, W), the facilitation of worker training (Senegal, W), and the establishment of mechanisms to enable ordinary workers to become familiar with ILO's work (Venezuela, W).
  4. Interagency cooperation and coordination: A number of replies suggested that the ILO's lead role in this area be asserted through more effective cooperation and coordination with other concerned international organizations such as the WTO, IMF, World Bank and the European Union (Russian Federation, G; San Marino, G; Slovakia, G; Senegal, W) to ensure, as far as possible, that workers' rights are not adversely affected by liberalization. Some constituents were of the view that the ILO should defend a formal linkage between respect for fundamental workers' rights and participation in international trade (workers from Argentina, Colombia, Italy and Sweden, among others). Some others, however, took the position that the ILO should not deal with trade matters (Malaysia, E) or seek to take a lead role or approval for the imposition of constraints (under the guise of protection) which may well do more harm than good (New Zealand, E). It was suggested that the ILO should call for a working party on the issue in the WTO (New Zealand, W) and obtain a standing position in the WTO (Finland, W). The creation of an ILO/WTO advisory committee on labour standards was also suggested (Czech Republic, W).

Geneva, 7 November 1996.

1. GB.265/WP/SDL/D.1, para. 2.

2. Copies of the questionnaire were sent to the ministers of labour of member States not present at the Conference in July 1996.

3. Replies were received from two different workers' organizations in eight member States. In order to avoid giving twice the weight to the workers' constituency from these countries, only one of the two replies could be included in the statistical results presented in this report. The selection was made according to the following criteria: where one of the replies was from the trade union confederation or congress, this was the one included (this was the case for Denmark, Pakistan, South Africa, Sri Lanka and Switzerland); in the case of France, where replies were received from two workers' confederations, the reply used was that from the confederation from which the Workers' delegate to the 1996 International Labour Conference (when the questionnaire was distributed) was appointed; in the two other cases the reply which arrived first was included (Argentina and Colombia). The statistical analysis is hence based on 128 replies from as many constituents. Despite their exclusion from the statistical analysis, however, the assessment parts of all 136 replies received are made use of in the report.

4. The regional grouping follows the ILO practice but as the ILO does not have an official breakdown of countries by the type of economy and level of development, this document relies on the classification used in the UNDP's Human Development Report 1996 to identify the least developed countries and the industrialized countries. The transition economies are those so identified in the World Bank's World Development Report 1996 and comprise the countries in Central and Eastern Europe, the newly independent States of the former Soviet Union, as well as China, Mongolia and Viet Nam (no replies were received from Mongolia).

5. It may be interesting to note that, when weighted by the number of constituents replying from each, the 76 countries represented in the sample account for about 31 per cent of the world's population, 44 per cent of its GDP, and 48 per cent of its total value of exports in 1993. Assessed in these terms, the sample may be regarded as somewhat more representative than suggested by the statistics in table 2.

6. Due to the small number of replies from the least developed countries, they are no longer distinguished as a separate category in this report.

7. Roughly one-fifth of the replies received did not include answers to the questions in this section. This rate of non-response is in fact typical of most questions in the questionnaire. Where the non-response rate is very different, it is pointed out in the text.

8. It should be emphasized that the scores, while useful as summary measures, are no substitute for percentage distributions and should be used in conjunction with them. The sign of the mean score is not meant to carry a value judgement.

9. A general caveat applies to comparisons involving overall perceptions or expectations of governments, employers, and workers. The observed differences may be influenced by the fact that the sets of countries that are implicitly involved in the comparison are usually not identical. For example, the workers that respond to a question may not come from the same countries as do governments that happen to respond to the same question. If so, the observed differences in views may not be due entirely to differences in perception about the same phenomenon. The bias due to country differences, however, can go in either direction. Furthermore, in most comparisons, there is indeed a good deal of country overlap, particularly in terms of the type of economy involved.

10. Two observations, however, may be in order in interpreting the evidence provided in this paragraph. The first is that the relevant question did not seek to associate any possible change in union membership rate with trade liberalization and globalization. While these processes may have had an effect on this rate, many other factors may also have been at work. For example, a decline in the unionization rate in some countries may be attributable mostly to a shift in the structure of employment, from the heavily unionized manufacturing industries to the less unionized services. Secondly, the substantial decline in the unionization rate in the transition economies need not imply a lower level of freedom of association compared to near-total membership in the earlier period.

11. In the case of Italy, the Government reply provided the figure of 71.2 per cent for all children of primary school age in 1981, although the corresponding figure for 1996 was 96.3 per cent. The employers' reply, however, put both at 100 per cent. For Greece, government estimates were 94.7 per cent in 1984-85 and 90.8 per cent in 1992-93 for all children aged 7-12. The figures for boys and girls were practically identical.

12. In the case of Hungary, however, while government figures were around 98 per cent for all primary school age children in both 1985 and 1995, the employers' reply put the corresponding ratios at 90 per cent in 1985 and 80 per cent in 1995 (the figures for boys and girls were identical).

13. It should be noted that the questions relating to school enrolment and violations of child labour were expected to be answered by governments only, but a number of employers' and workers' organizations also answered these questions and their answers were not always identical to those of their governments.

14. An African regional workshop on the protection of workers' rights and working conditions in export processing zones was held in Johannesburg from 15-18 July 1996 under the auspices of the ILO.

15. The questionnaire defined fundamental workers' rights as freedom of association and the right to engage in collective bargaining, and protection against forced labour, child labour and discrimination (including equal remuneration for work of equal value).


(TABLES: TRAITEXT doc. GB267\11-11.E96)

Updated by VC. Approved by NdW. Last update: 26 January 2000.